Deprecate 2x Tensorflow, Keras and ONNX (#2199)

Signed-off-by: Xin He <[email protected]>

Author: xin3he

docs/source/adaptor.md

@@ -24,9 +24,9 @@ tuning strategy and vanilla framework quantization APIs.
 
 |Framework     |Adaptor      |
 |--------------|:-----------:|
-|TensorFlow    |✔     |
+|TensorFlow (Deprecated)    |✔     |
 |PyTorch       |✔     |
-|ONNX          |✔     |
+|ONNX (Deprecated)          |✔     |
 
 
 ## Working Flow

docs/source/benchmark.md

@@ -1,5 +1,7 @@
-Benchmarking
-============
+Benchmarking (Deprecated)
+===============
+> Since version >= 3.3, this functionality has been replaced by [incbench](https://github.com/intel/neural-compressor/blob/master/docs/source/3x/benchmark.md), which is easier to use.
+----------------------------
 1. [Introduction](#Introduction)
 2. [Benchmark Support Matrix](#Benchmark-Support-Matrix)
 3. [Get Started with Benchmark](#Get-Started-with-Benchmark)

docs/source/dataloader.md

@@ -31,10 +31,10 @@ Of cause, users can also use frameworks own dataloader in Neural Compressor.
 
 | Framework     | Status     |
 |---------------|:----------:|
-| TensorFlow    |  ✔  |
-| Keras         |  ✔  |
+| TensorFlow (Deprecated)    |  ✔  |
+| Keras (Deprecated)         |  ✔  |
 | PyTorch       |  ✔  |
-| ONNX Runtime   |  ✔  |
+| ONNX Runtime (Deprecated)   |  ✔  |
 
 ## Get Started with DataLoader
 

docs/source/distillation_quantization.md

@@ -23,7 +23,7 @@ Intel provides a quantization-aware training (QAT) method that incorporates a no
 
 ### Distillation for Quantization Support Matrix
 
-|**Algorithm**                      |**PyTorch**   |**TensorFlow** |
+|**Algorithm**                      |**PyTorch**   |**TensorFlow (Deprecated)** |
 |---------------------------------|:--------:|:---------:|
 |Distillation for Quantization    |✔  |✖   |
 

docs/source/distributed.md

@@ -16,7 +16,7 @@ Distributed Training and Inference (Evaluation)
 
     4.1. [Pytorch Examples](#pytorch-examples)
 
-    4.2. [Tensorflow Examples](#tensorflow-examples)
+    4.2. [Tensorflow Examples](#tensorflow-examples) (Deprecated)
 ## Introduction
 
 Neural Compressor uses [horovod](https://github.com/horovod/horovod) for distributed training.
@@ -33,8 +33,8 @@ Distributed training and inference are supported in PyTorch and TensorFlow curre
 |------------|---------|:-------------------:|
 | PyTorch    | QAT     |       ✔      |
 | PyTorch    | PTQ     |       ✔      |
-| TensorFlow | PTQ     |       ✔      |
-| Keras      | Pruning |       ✔      |
+| TensorFlow (Deprecated) | PTQ     |       ✔      |
+| Keras (Deprecated)      | Pruning |       ✔      |
 
 ## Get Started with Distributed Training and Inference API
 To enable distributed training or inference, the steps are:
@@ -168,7 +168,7 @@ Horovodrun requires user set up SSH on all hosts without any prompts. To do dist
 - PyTorch example-2: QAT (Quantization Aware Training)
   - Please follow this README.md exactly：[QAT](../../examples/pytorch/image_recognition/torchvision_models/quantization/qat/eager/distributed)
 
-### TensorFlow Examples:
+### TensorFlow Examples: (Deprecated)
 - TensorFlow example-1: 'ResNet50 V1.0' PTQ (Post Training Quantization) with distributed inference    
   - Step-1: Please cd (change directory) to the [TensorFlow Image Recognition Example](../../examples/tensorflow/image_recognition) and follow the readme to run PTQ, ensure that PTQ of 'ResNet50 V1.0' can be successfully executed.
   - Step-2: We only need to modify the [resnet50_v1.yaml](../../examples/tensorflow/image_recognition/tensorflow_models/quantization/ptq/resnet50_v1.yaml), add a line 'distributed: True' in the 'evaluation' field.

docs/source/export.md

@@ -48,7 +48,7 @@ Here is the workflow of our export API for PyTorch/Tensorflow FP32/INT8 model.
     <td>QOperator/QDQ INT8</td>
   </tr>
   <tr>
-    <td rowspan="3">TensorFlow</td>
+    <td rowspan="3">TensorFlow (Deprecated)</td>
     <td>FP32</td>
     <td>FP32</td>
   </tr>

docs/source/framework_yaml.md

@@ -25,9 +25,9 @@ unless they have clear requirements that can only be met by modifying the attrib
 
 | Framework  | YAML Configuration Files |
 |------------|:------------------------:|
-| TensorFlow |         ✔         |
+| TensorFlow (Deprecated) |         ✔         |
 | PyTorch    |         ✔         |
-| ONNX       |         ✔         |
+| ONNX (Deprecated)       |         ✔         |
 
 
 ## Get started with Framework YAML Files

docs/source/infrastructure.md

@@ -34,11 +34,11 @@ Intel® Neural Compressor has unified interfaces which dispatch tasks to differe
             <td align="center"><a href="https://pytorch.org/docs/stable/quantization.html#eager-mode-quantization">PyTorch Eager</a>/<a href="https://pytorch.org/docs/stable/quantization.html#prototype-fx-graph-mode-quantization">PyTorch FX</a>/<a href="https://github.com/intel/intel-extension-for-pytorch">IPEX</a></td>
         </tr>
         <tr>
-            <td align="center">TensorFlow</td>
+            <td align="center">TensorFlow (Deprecated)</td>
             <td align="center"><a href="https://github.com/tensorflow/tensorflow">TensorFlow</a>/<a href="https://github.com/Intel-tensorflow/tensorflow">Intel TensorFlow</a></td>
         </tr>
         <tr>
-            <td align="center">ONNX Runtime</td>
+            <td align="center">ONNX Runtime (Deprecated)</td>
             <td align="center"><a href="https://github.com/microsoft/onnxruntime/blob/master/onnxruntime/python/tools/quantization/quantize.py">QLinearops/QDQ</a></td>
         </tr>
         <tr>
@@ -47,7 +47,7 @@ Intel® Neural Compressor has unified interfaces which dispatch tasks to differe
             <td align="center"><a href="https://pytorch.org/docs/stable/quantization.html#eager-mode-quantization">PyTorch eager mode</a>/<a href="https://pytorch.org/docs/stable/quantization.html#prototype-fx-graph-mode-quantization">PyTorch fx mode</a>/<a href="https://github.com/intel/intel-extension-for-pytorch">IPEX</a></td>
         </tr>
         <tr>
-            <td align="center">ONNX Runtime</td>
+            <td align="center">ONNX Runtime (Deprecated)</td>
             <td align="center"><a href="https://github.com/microsoft/onnxruntime/blob/master/onnxruntime/python/tools/quantization/quantize.py">QIntegerops</a></td>
         </tr>  
         <tr>
@@ -56,7 +56,7 @@ Intel® Neural Compressor has unified interfaces which dispatch tasks to differe
             <td align="center"><a href="https://pytorch.org/docs/stable/quantization.html#eager-mode-quantization">PyTorch eager mode</a>/<a href="https://pytorch.org/docs/stable/quantization.html#prototype-fx-graph-mode-quantization">PyTorch fx mode</a>/<a href="https://github.com/intel/intel-extension-for-pytorch">IPEX</a></td>
         </tr>
         <tr>
-            <td align="center">TensorFlow</td>
+            <td align="center">TensorFlow (Deprecated)</td>
             <td align="center"><a href="https://github.com/tensorflow/tensorflow">TensorFlow</a>/<a href="https://github.com/Intel-tensorflow/tensorflow">Intel TensorFlow</a></td>
         </tr>
     </tbody>
@@ -79,7 +79,7 @@ Intel® Neural Compressor has unified interfaces which dispatch tasks to differe
     <td rowspan="3">Unstructured Pruning</td>
     <td rowspan="3">Element-wise</td>
     <td>Magnitude</td>
-    <td>PyTorch, TensorFlow</td>
+    <td>PyTorch, TensorFlow (Deprecated)</td>
   </tr>
   <tr>
     <td>Pattern Lock</td>
@@ -185,6 +185,6 @@ Intel® Neural Compressor has unified interfaces which dispatch tasks to differe
 [Mixed precision](mixed_precision.md):
 |Framework     |         |
 |--------------|:-----------:|
-|TensorFlow    |&#10004;     |
+|TensorFlow (Deprecated)    |&#10004;     |
 |PyTorch       |&#10004;     |
-|ONNX          |plan to support in the future |
+|ONNX (Deprecated)          |plan to support in the future |

docs/source/metric.md

@@ -3,11 +3,11 @@ Metrics
 1. [Introduction](#introduction)
 2. [Supported Built-in Metric Matrix](#supported-built-in-metric-matrix)
 
-    2.1. [TensorFlow](#tensorflow)
+    2.1. [TensorFlow (Deprecated)](#tensorflow)
 
     2.2. [PyTorch](#pytorch)
 
-    2.3. [ONNXRT](#onnxrt)
+    2.3. [ONNXRT (Deprecated)](#onnxrt)
 
 3. [Get Started with Metric](#get-started-with-metric)
 

docs/source/mixed_precision.md

@@ -49,7 +49,7 @@ The recently launched 3rd Gen Intel® Xeon® Scalable processor (codenamed Coope
             <td align="left">:x:</td>
         </tr>
         <tr>
-            <td rowspan="4" align="left">ONNX Runtime</td>
+            <td rowspan="4" align="left">ONNX Runtime (Deprecated)</td>
             <td align="left">CPUExecutionProvider</td>
             <td align="left">MLAS</td>
             <td align="left">"default"</td>
@@ -82,7 +82,7 @@ The recently launched 3rd Gen Intel® Xeon® Scalable processor (codenamed Coope
             <td align="left">:x:</td>
         </tr>
         <tr>
-            <td rowspan="2" align="left">Tensorflow</td>
+            <td rowspan="2" align="left">Tensorflow (Deprecated)</td>
             <td align="left">Tensorflow</td>
             <td align="left">OneDNN</td>
             <td align="left">"default"</td>

docs/source/quantization_mixed_precision.md

@@ -15,7 +15,7 @@ q_model = quantization.fit(
 )
 ```
 
-### Tensorflow
+### Tensorflow (Deprecated)
 
 Intel has worked with the TensorFlow development team to enhance TensorFlow to include bfloat16 data support for CPUs. For more information about BF16 in TensorFlow, please read [Accelerating AI performance on 3rd Gen Intel® Xeon® Scalable processors with TensorFlow and Bfloat16](https://blog.tensorflow.org/2020/06/accelerating-ai-performance-on-3rd-gen-processors-with-tensorflow-bfloat16.html).
 

examples/README.md

@@ -2,25 +2,28 @@ Examples
 ==========
 Intel® Neural Compressor validated examples with multiple compression techniques, including quantization, pruning, knowledge distillation and orchestration. Part of the validated cases can be found in the example tables, and the release data is available [here](../docs/source/validated_model_list.md).
 
+> Note: `3.x_api` here is the target example folder for version >= 3.0, which contains pytorch and tensorflow related examples. TensorFlow and ONNX related examples here are no longer maintained. 
+
+
 # Quick Get Started Notebook Examples
-* [Quick Get Started Notebook of Intel® Neural Compressor for ONNXRuntime](/examples/notebook/onnxruntime/Quick_Started_Notebook_of_INC_for_ONNXRuntime.ipynb)
+* [Quick Get Started Notebook of Intel® Neural Compressor for ONNXRuntime (Deprecated)](/examples/notebook/onnxruntime/Quick_Started_Notebook_of_INC_for_ONNXRuntime.ipynb)
 
-* [Quick Get Started Notebook of Intel® Neural Compressor for Tensorflow](/examples/notebook/tensorflow/resnet/resnet_quantization.ipynb)
+* [Quick Get Started Notebook of Intel® Neural Compressor for Tensorflow (Deprecated)](/examples/notebook/tensorflow/resnet/resnet_quantization.ipynb)
 
 * [Quick Get Started Notebook of Intel® Neural Compressor for Pytorch](/examples/notebook/pytorch/Quick_Started_Notebook_of_INC_for_Pytorch.ipynb)
 
 # Helloworld Examples
 * [torch_llm](/examples/helloworld/torch_llm): apply the weight-only quantization to LLMs.
 * [torch_non_llm](/examples/helloworld/torch_non_llm): apply the static quantization to non-LLMs.
-* [tf_example1](/examples/helloworld/tf_example1): quantize with built-in dataloader and metric.
-* [tf_example2](/examples/helloworld/tf_example2): quantize keras model with customized metric and dataloader.
-* [tf_example3](/examples/helloworld/tf_example3): convert model with mix precision.
-* [tf_example4](/examples/helloworld/tf_example4): quantize checkpoint with dummy dataloader.
-* [tf_example5](/examples/helloworld/tf_example5): config performance and accuracy measurement.
-* [tf_example6](/examples/helloworld/tf_example6): use default user-facing APIs to quantize a pb model.
-* [tf_example7](/examples/helloworld/tf_example7): quantize and benchmark with pure python API.
+* [tf_example1 (Deprecated)](/examples/helloworld/tf_example1): quantize with built-in dataloader and metric.
+* [tf_example2 (Deprecated)](/examples/helloworld/tf_example2): quantize keras model with customized metric and dataloader.
+* [tf_example3 (Deprecated)](/examples/helloworld/tf_example3): convert model with mix precision.
+* [tf_example4 (Deprecated)](/examples/helloworld/tf_example4): quantize checkpoint with dummy dataloader.
+* [tf_example5 (Deprecated)](/examples/helloworld/tf_example5): config performance and accuracy measurement.
+* [tf_example6 (Deprecated)](/examples/helloworld/tf_example6): use default user-facing APIs to quantize a pb model.
+* [tf_example7 (Deprecated)](/examples/helloworld/tf_example7): quantize and benchmark with pure python API.
 
-# TensorFlow Examples
+# TensorFlow Examples (Deprecated)
 ## Quantization
 <table>
 <thead>
@@ -970,7 +973,7 @@ Intel® Neural Compressor validated examples with multiple compression technique
 </tbody>
 </table>
 
-# ONNX Runtime Examples
+# ONNX Runtime Examples (Deprecated)
 ## Quantization
 
 <table>
@@ -1472,7 +1475,7 @@ Intel® Neural Compressor validated examples with multiple compression technique
 
 # Notebook Examples
 
-* [Performance of FP32 Vs. INT8 ResNet50 Model](/examples/notebook/perf_fp32_int8_tf): compare existed FP32 & INT8 ResNet50 model directly.
-* [Intel® Neural Compressor Sample for TensorFlow*](/examples/notebook/tensorflow/alexnet_mnist): an End-To-End pipeline to build up a CNN model by TensorFlow to recognize handwriting number and speed up AI model by Intel® Neural Compressor.
-* [Accelerate VGG19 Inference on Intel® Gen4 Xeon® Sapphire Rapids](/examples/notebook/tensorflow/vgg19_ibean): an End-To-End pipeline to train VGG19 model by transfer learning based on pre-trained model from [TensorFlow Hub](https://tfhub.dev); quantize it by Intel® Neural Compressor on Intel® Gen4 Xeon® Sapphire Rapids.
+* [Performance of FP32 Vs. INT8 ResNet50 Model (Deprecated)](/examples/notebook/perf_fp32_int8_tf): compare existed FP32 & INT8 ResNet50 model directly.
+* [Intel® Neural Compressor Sample for TensorFlow* (Deprecated)](/examples/notebook/tensorflow/alexnet_mnist): an End-To-End pipeline to build up a CNN model by TensorFlow to recognize handwriting number and speed up AI model by Intel® Neural Compressor.
+* [Accelerate VGG19 Inference on Intel® Gen4 Xeon® Sapphire Rapids (Deprecated)](/examples/notebook/tensorflow/vgg19_ibean): an End-To-End pipeline to train VGG19 model by transfer learning based on pre-trained model from [TensorFlow Hub](https://tfhub.dev); quantize it by Intel® Neural Compressor on Intel® Gen4 Xeon® Sapphire Rapids.
 

examples/helloworld/README.md

@@ -5,7 +5,7 @@
 * [torch_woq](/examples/helloworld/torch_woq): apply the weight-only quantization to LLMs.
 * [torch_static_quant](/examples/helloworld/torch_static_quant): apply the static quantization to non-LLMs.
 
-## Tensorflow Examples
+## Tensorflow Examples (Deprecated)
 ### Prerequisite
 Enter the following commands to prepare a dataset and pretrained models for the included Hello World examples:
 

examples/helloworld/tf_example1/README.md

@@ -1,4 +1,4 @@
-tf_example1 example
+tf_example1 example (Deprecated)
 =====================
 
 Step-by-Step

examples/helloworld/tf_example2/README.md

@@ -1,4 +1,4 @@
-tf_example2 example
+tf_example2 example (Deprecated)
 =====================
 
 Step-by-Step

examples/helloworld/tf_example3/README.md

@@ -1,4 +1,4 @@
-tf_example3 example
+tf_example3 example (Deprecated)
 =====================
 
 Step-by-Step

examples/helloworld/tf_example4/README.md

@@ -1,4 +1,4 @@
-tf_example4 example
+tf_example4 example (Deprecated)
 =====================
 
 Step-by-Step

examples/helloworld/tf_example5/README.md

@@ -1,4 +1,4 @@
-tf_example5 example
+tf_example5 example (Deprecated)
 =====================
 
 Step-by-Step

examples/helloworld/tf_example6/README.md

@@ -1,4 +1,4 @@
-tf_example6 example
+tf_example6 example (Deprecated)
 =====================
 
 Step-by-Step

examples/helloworld/tf_example7/README.md

@@ -1,4 +1,4 @@
-tf_example7 example
+tf_example7 example (Deprecated)
 =====================
 This example is used to demonstrate how to quantize a TensorFlow model with dummy dataset.